ModelPreset

The ModelPreset class defines the preset configuration for AI models.

Properties

• model (str): Name of the AI model (e.g. gpt-3.5-turbo)
• name (str): Identifier name for current preset, defaults to "default"
• base_url (str): Base address of API service (uses OpenAI default if empty)
• api_key (str): Key required to access API
• protocol (str): Protocol adapter type, defaults to "main"
• config (ModelConfig): Model configuration object
• thinking_config (ThinkingConfig | None): Thinking/reasoning configuration for models that support it (e.g., OpenAI o1, Anthropic extended thinking)
• rate (float | None): Token cost rate for the model, used for cost estimation
• extra (dict[str, Any]): Extra configuration items

Methods

• load(path: Path): Load model preset configuration from the specified path
• save(path: Path): Save current preset configuration to the specified path

Description

The ModelPreset class inherits from BaseModel and is used to encapsulate complete configuration information for AI models. It not only includes basic model parameters but also provides functionality to load and save configuration files, facilitating management and reuse of different model configurations.